Extreme Learning Machines for Attention-based Multiple Instance Learning in Whole-Slide Image Classification

2025-03-13

AI Summary

Detecting rare circulating cells (e.g., normoblasts) in whole-slide images faces challenges of scarce instance-level annotations and computationally inefficient models. Method: We propose a lightweight attention-enhanced multiple-instance learning (MIL) framework that integrates extreme learning machines (ELMs) with a learnable attention mechanism to construct an end-to-end classifier; additionally, we introduce a high-dimensional nonlinear feature mapping strategy to improve robustness and generalization. Results: Experiments show that our method reduces trainable parameters by 5× while incurring only a 1.5% AUC drop; it outperforms baseline MIL models by over 10% in AUC. Ablation studies confirm the critical contribution of the nonlinear module—its removal degrades AUC by more than 4%. This work establishes a novel paradigm for accurate and efficient rare-cell identification in computational pathology.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Multi-modal Vision
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Whole-slide image classification represents a key challenge in computational pathology and medicine. Attention-based multiple instance learning (MIL) has emerged as an effective approach for this problem. However, the effect of attention mechanism architecture on model performance is not well-documented for biomedical imagery. In this work, we compare different methods and implementations of MIL, including deep learning variants. We introduce a new method using higher-dimensional feature spaces for deep MIL. We also develop a novel algorithm for whole-slide image classification where extreme machine learning is combined with attention-based MIL to improve sensitivity and reduce training complexity. We apply our algorithms to the problem of detecting circulating rare cells (CRCs), such as erythroblasts, in peripheral blood. Our results indicate that nonlinearities play a key role in the classification, as removing them leads to a sharp decrease in stability in addition to a decrease in average area under the curve (AUC) of over 4%. We also demonstrate a considerable increase in robustness of the model with improvements of over 10% in average AUC when higher-dimensional feature spaces are leveraged. In addition, we show that extreme learning machines can offer clear improvements in terms of training efficiency by reducing the number of trained parameters by a factor of 5 whilst still maintaining the average AUC to within 1.5% of the deep MIL model. Finally, we discuss options of enriching the classical computing framework with quantum algorithms in the future. This work can thus help pave the way towards more accurate and efficient single-cell diagnostics, one of the building blocks of precision medicine.

Problem

• Improving whole-slide image classification using attention-based MIL.
• Enhancing model sensitivity and reducing training complexity with extreme learning machines.
• Exploring higher-dimensional feature spaces for better classification performance.

Innovation

• Combines extreme learning with attention-based MIL
• Uses higher-dimensional feature spaces for deep MIL
• Reduces training parameters while maintaining AUC